User profiles for G. Brostow
Gabriel J. BrostowProfessor of Computer Science, University College London Verified email at cs.ucl.ac.uk Cited by 15139 |
Unsupervised monocular depth estimation with left-right consistency
…, O Mac Aodha, GJ Brostow - Proceedings of the IEEE …, 2017 - openaccess.thecvf.com
Learning based methods have shown very promising results for the task of depth estimation
in single images. However, most existing approaches treat depth prediction as a supervised …
in single images. However, most existing approaches treat depth prediction as a supervised …
Digging into self-supervised monocular depth estimation
…, M Firman, GJ Brostow - Proceedings of the …, 2019 - openaccess.thecvf.com
… We additionally show we can get higher scores at a larger 1024 × 320 resolution, similar to
[47] – see supplementary Section G. These high resolution numbers are bolded if they beat …
[47] – see supplementary Section G. These high resolution numbers are bolded if they beat …
Semantic object classes in video: A high-definition ground truth database
Visual object analysis researchers are increasingly experimenting with video, because it is
expected that motion cues should help with detection, recognition, and other analysis tasks. …
expected that motion cues should help with detection, recognition, and other analysis tasks. …
Segmentation and recognition using structure from motion point clouds
We propose an algorithm for semantic segmentation based on 3D point clouds derived from
ego-motion. We motivate five simple cues designed to model specific patterns of motion and …
ego-motion. We motivate five simple cues designed to model specific patterns of motion and …
Harmonic networks: Deep translation and rotation equivariance
…, D Turmukhambetov, GJ Brostow - Proceedings of the …, 2017 - openaccess.thecvf.com
Translating or rotating an input image should not affect the results of many computer vision
tasks. Convolutional neural networks (CNNs) are already translation equivariant: input image …
tasks. Convolutional neural networks (CNNs) are already translation equivariant: input image …
Unsupervised bayesian detection of independent motion in crowds
GJ Brostow, R Cipolla - 2006 IEEE computer society …, 2006 - ieeexplore.ieee.org
While crowds of various subjects may offer applicationspecific cues to detect individuals, we
demonstrate that for the general case, motion itself contains more information than …
demonstrate that for the general case, motion itself contains more information than …
The temporal opportunist: Self-supervised multi-frame monocular depth
…, V Prisacariu, G Brostow… - Proceedings of the …, 2021 - openaccess.thecvf.com
… [23] Clément Godard, Oisin Mac Aodha, and Gabriel J Brostow. Unsupervised monocular
depth … Brostow. Digging into self-supervised monocular depth estimation. In ICCV, 2019. …
depth … Brostow. Digging into self-supervised monocular depth estimation. In ICCV, 2019. …
Deep blending for free-viewpoint image-based rendering
Free-viewpoint image-based rendering (IBR) is a standing challenge. IBR methods combine
warped versions of input photos to synthesize a novel view. The image quality of this …
warped versions of input photos to synthesize a novel view. The image quality of this …
[HTML][HTML] Bat detective—Deep learning tools for bat acoustic signal detection
…, E Tilova, M Girolami, G Brostow… - PLoS computational …, 2018 - journals.plos.org
Passive acoustic sensing has emerged as a powerful tool for quantifying anthropogenic
impacts on biodiversity, especially for echolocating bat species. To better assess bat population …
impacts on biodiversity, especially for echolocating bat species. To better assess bat population …
Self-supervised monocular depth hints
Monocular depth estimators can be trained with various forms of self-supervision from
binocular-stereo data to circumvent the need for high-quality laser-scans or other ground-truth data…
binocular-stereo data to circumvent the need for high-quality laser-scans or other ground-truth data…